Articles

Access the latest knowledge in applied science, electrical engineering, computer science and information technology, education, and health.

Advanced Search
Filter Icon

Filters article

Years

FAQ Arrow
Single Year
Range Year
0
0

Source Title

FAQ Arrow

Authors

FAQ Arrow

29,922 Article Results

Interrogative insights into depression detection via social networks and machine learning techniques

10.11591/ijeecs.v40.i1.pp388-396
Chaithra Indavara Venkateshagowda , Roopashree Hejjajji Ranganathasharma , Yogeesh Ambalagere Chandrashekaraiah
As users on social networks (SNs) interact with one another by exchanging information, giving feedback, finding new content, and participating in discussions; thus, generating large volumes of data each day. This data includes images, texts, videos and can be used to help the user find out how they have been doing, when they were depressed, how not to be depressed, and other similar insights. Depression is one of the most common chronic illnesses and it has emerged as a global mental health problem. But the lack of these data is incomplete, sparse and sometimes inaccurate, and so the task of diagnosing depression using automated systems is still proving a challenge. Various techniques have been used to detect depression through the years however, machine learning (ML) and deep learning (DL) techniques offer better ways. In the context of that, this study reviews state-of-the-art ML and DL approaches for the detection of depression using systematic literature review (SLR) method as well as highlight fundamental challenges in literature, which future works can focus on. We hope that this survey will provide a better understanding of these strategies for the readers and researchers in the ML and DL fields, when it comes to diagnosis of depression.
Volume: 40
Issue: 1
Page: 388-396
Publish at: 2025-10-01

Autoencoder-based Gaussian mixture model for diagnosing early onset of diabetic retinopathy

10.11591/ijeecs.v40.i1.pp164-172
Priyanka Sreenivas , Kavita V. Horadi , Kalpa Rajashekar
The current study presents a simplified yet innovative solution towards effective early diagnosis of diabetic retinopathy (DR) that leads to irreversible blindness. A review of current literature shows a considerable number of machine learning and deep learning approaches have been presented; however, there are significant issues with the early detection of DR. Hence, the proposed study deploys a novel architecture using an autoencoder that extracts a hidden representation of retinal images while binary classification is carried out using a Gaussian mixture model. The prime contribution is the joint integration of deep learning with statistical modelling towards efficient feature extraction and anomaly detection, supporting early determination of DR. The study outcome shows a proposed system to significantly exhibit 96.5% accuracy, 94.2% sensitivity, and 98.3% specificity on two standard benchmarked datasets in comparison to existing models frequently used for the diagnosis of DR.
Volume: 40
Issue: 1
Page: 164-172
Publish at: 2025-10-01

Unveiling the influence of back-translation on sentiment analysis of Indonesian hotel reviews

10.11591/ijeecs.v40.i1.pp271-279
Sandy Kurniawan , Retno Kusumaningrum , Priyo Sidik Sasongko
This study aims to conduct sentiment analysis on hotel reviews in Indonesian using several machine learning classification algorithms, namely multinomial naive bayes (MNB), support vector machine (SVM), and random forest (RF). The back translation method is employed to generate synthetic data variations that are used as additional data variations in building classification models. This research tests three scenarios based on the datasets used: the original dataset, the dataset resulting from back translation, and the combined dataset of both. The experimental results show that the use of combined data yields better results, with the random forest algorithm standing out as the best performer. Back translation significantly improves model evaluation in sentiment analysis for several reasons, including enriching the dataset with new variations, enhancing model robustness, and increasing dataset complexity. However, the differences in the number of word features among scenarios indicate that back translation also significantly influences the dataset's characteristics.
Volume: 40
Issue: 1
Page: 271-279
Publish at: 2025-10-01

Analysis of solid oxide fuel cell systems for off-grid energy production

10.11591/ijeecs.v40.i1.pp18-33
Boudjemaa Mehimmedetsi , Abdellah Draidi , Billel Smaani
This work presents a simulation study of a 50-kW solid oxide fuel cell (SOFC) power supply system that provides electricity to residential users. Indeed, many decentralized applications rely on renewable energy sources not connected to the primary power grid. Moreover, fuel cell modelling and simulation are critical for promoting renewable energy as they eliminate the need for physical prototypes, saving time and money. We have also developed a reliable model for simulating self-contained SOFC fuel cells. The elaborated model includes the kinetics of electrochemical processes and accounts for voltage losses in SOFCs. Our fuel cells produce the necessary electrical current to charge the device. Also, our system has fuel cells, a DC/DC converter, and an inverter with LCL filters. These components connect the fuel cell system to other power electronics and the electrical load. Furthermore, a mathematical model of a dual inverter system describes its control method, including the proportional and integral parameters in the voltage and current loops has been derived. The proposed model and system could be helpful for a standalone load.
Volume: 40
Issue: 1
Page: 18-33
Publish at: 2025-10-01

An improved conversation emotion detection using hybrid f-nn classifier

10.11591/ijeecs.v40.i1.pp490-498
Abhishek A. Vichare , Satishkumar L. Varma
Emotion recognition from text is a crucial task in natural language processing (NLP) with applications in sentiment analysis, human-computer interaction, and psychological research. In this study, we present a novel approach for text-based emotion recognition using a modified firefly algorithm (MFA). The firefly algorithm is a swarm intelligence method inspired by the bioluminescent communication of fireflies, and it is known for its simplicity and efficiency in optimization tasks. In this paper MFAbased model is evaluated on the international survey on emotion antecedents and reactions (ISEAR) dataset, which includes text entries categorized by various emotions. Experimental results indicate that our approach achieved promising outcomes. Specifically, the proposed method, which combines the firefly algorithm with a multilayer perceptron (MLP), attained an accuracy of 92.07%, surpassing most other approaches reported in the literature.
Volume: 40
Issue: 1
Page: 490-498
Publish at: 2025-10-01

Predict glucose values with DE algorithm optimized T-LSTM

10.11591/ijeecs.v40.i1.pp530-544
QingXiang Bian , Azizan As’array , XiangGuo Cong , Khairil Anas bin Md Rezali , Raja Mohd Kamil bin Raja Ahmad , Mohd Zarhamdy Md. Zain
The prevalence of diabetes is rising. According to the International Diabetes Federation (IDF) predictions, the number of diabetic patients worldwide will reach 608 million in 2030, accounting for approximately 11.3% of the total number of people in the world. To monitor and predict the future 1 hour glucose have a great significance meaning for patients. This research utilizes a differential evolution (DE) algorithm, an optimized hybrid model transformer and long short-term memory (T-LSTM) technologies to analyze historical data from continuous blood glucose monitoring (CGM) systems and equipment calibration values. The aim is to predict future blood sugar levels in patients, thereby helping to prevent episodes of hypoglycemia and hyperglycemia. The study tested the model using the CGM data from 8 patients at the Suzhou Municipal Hospital in Jiangsu Province, China. Results show that this DE-optimized T-LSTM model outperforms traditional models. The model's accuracy is evaluated using mean squared error (MSE), with MSE values recorded at 15, 30, and 45 minutes being 0.96, 1.54, and 2.31, respectively.
Volume: 40
Issue: 1
Page: 530-544
Publish at: 2025-10-01

MoBiSafe: an obfuscated single factor authentication mode to enhance secured USSD channel transaction in Nigeria

10.11591/ijeecs.v40.i1.pp426-436
Amaka Patience Binitie , Sebastina Nkechi Okofu , Margaret Dumebi Okpor , Kizito Eluemunor Anazia , Arnold Adimabua Ojugo , Francesca Avwuru Egbokhare , Annie Egwali , Peace Oguguo Ezzeh , Rita Erhovwo Ako , Victor Ochuko Geteloma , Tabitha Chukwudi Aghaunor , Eferhire Valentine Ugbotu , Sunny Innocent Onyemenem
The flexibility of the unstructured supplementary service data (USSD) across mobile phones has caused its adoption surge as a payment channel. Its usage accommodates financial inclusivity and extends customer reach irrespective of their specific phone capabilities. With data conveyed on the USSD channel in plaintext–this has raised vulnerability issues with shoulder surfing attacks. The use of password yielded extra layer of security as authentication to USSD-based services. But, the rise in password guess attacks has necessitated a new scheme. This study is a randomized-obfuscated single factor authentication (SFA) mode via a 5-digit PIN-entry as requisite for the USSD channel. It yields a list via which users select a key-array that corresponds to their PIN as concealed in a 10-digit array. Expert assess of MoBiSafe’s usability and security against shoulder-surf yielded 10.1 msecs and 2.26 msecs respectively to outperform existing models that utilize direct/indirect PIN-entry as in USSD transactions. And this was found to be both secure, usable and acceptable.
Volume: 40
Issue: 1
Page: 426-436
Publish at: 2025-10-01

Computer vision based smart overspeeding vehicle surveillance system

10.11591/ijece.v15i5.pp4740-4750
Budhaditya Bhattacharjee , Pragyendra Pragyendra , Boopalan Ganapathy , Shanmugasundaram M.
In India, overspeeding causes more than 60% of deaths. Therefore, we need a system that tracks the median speed of cars and identifies those who regularly violate the law. Road fatalities can be reduced as a result of maintaining law and order. In this paper, we present an embedded system that can read the license plates of passing cars in real time. Using optical character recognition technology, the proposed system will capture images of license plates. In addition, it sends short message service (SMS) notifications regarding the highway speed of a specific moving vehicle violating the rules to the relevant authorities. By using this technique, several manual operations that were previously required to detect over-speeding automobiles with RADAR guns are eliminated. On the roadway, the device can only be operated by one operator due to its well-developed user interface. As part of this work, a downloadable database is developed which includes information about speeding vehicles as well as vehicles travelling on a roadway at the moment they are detected.
Volume: 15
Issue: 5
Page: 4740-4750
Publish at: 2025-10-01

Inverse-Mel scale spectrograms for high-frequency feature extraction and audio anomaly detection in industrial machines

10.11591/ijai.v14.i5.pp3656-3666
Kader Basha Tajuddin Shaikh , Naresh P. Jawarkar , Vasif Ahmed , Nadir Nizar Ali Charniya
Unlike humans, the energies in industrial machine sounds (IMS) vary across a wide range of frequencies. Mel scales, which are developed for the perception of human audio, fail to capture the complete information present in IMS. To improve performance, we propose using an inverse-Mel scale, along with the concatenation and combination of Mel and inverse-Mel scale based spectrograms, as feature vectors for audio anomaly detection (AAD) in industrial machines. Adaptation in the Librosa Python package and the DCASE 2022 Challenge Task 2 baseline system is pursued for the construction of inverse-Mel scale spectrograms. Experiments are conducted using the malfunctioning industrial machine investigation and inspection for domain generalization (MIMII DG) datasets. Systems based on the inverse-Mel scale achieve a maximum improvement of up to 37% in the bearing machine and an average improvement of up to 9% in the area under the curve (AUC) score across all machines in the MIMII DG datasets. The proposed features also enhance DG, overcoming the effects of environmental and operational domain shifts caused by variations in recording setup, load, background noise, and operational patterns. Challenge official evaluator assessed the proposed system against the evaluation datasets, ranking it three positions higher than the baseline system.
Volume: 14
Issue: 5
Page: 3656-3666
Publish at: 2025-10-01

Wind farm integration with the objective of transmission expansion power in South Africa

10.11591/ijeecs.v40.i1.pp34-46
Nomihla Wandile Ndlela , Katleho Moloi , Musasa Kabeya
Growing renewable energy (RE) use mitigates climate change. The integration of large-scale intermittent renewable energy resources (RER) like wind energy into electrical networks has increased during the past decade. However, careful planning is needed to accommodate the long-term energy demand increase. Transmission network expansion planning (TNEP) is the methodical and profitable process of increasing power infrastructure to meet predicted electricity demand while preserving reliability. This article is for those interested in integrating renewable energy sources (RES) into HVTL to increase power availability and decrease losses. The Eros-VuyaniNeptune 400 kV transmission powerline connecting KwaZulu-Natal to the Eastern Cape is used in this study. It was implemented during the transfer of affected residents in the Ingquza Hill Local Municipality, which includes Lusikisiki and Flagstaff villages. This study connects the existing Metro wind farm to the Vuyani substation, which is connected to the Eros substation through a 400 kV transmission line. This research enhanced transmission line power while preserving grid stability with a 27 MW wind farm, and also increased external grid reserve capacity for future usage or unexpected power demand. This paper outlines TNEP’s significant advances using classic (mathematical) and advanced (heuristic and meta-heuristic) optimization approaches.
Volume: 40
Issue: 1
Page: 34-46
Publish at: 2025-10-01

Path planning of an elongated undulating fin using mutant particle swarm optimization

10.11591/ijeecs.v40.i1.pp10-17
Thi Thom Hoang , Thi Huong Le
This paper proposes a mutant particle swarm optimization algorithm (M-PSO) to optimize the power energy of a bio-mimetic robotic fish that comprises sixteen undulating fin-rays equipped to a fish robot. The main objective is to obtain the shortest path for the fish robot to achieve the desired position while minimizing power consumption. The proposed MPSO is a recent generation of particle swarm optimization (PSO) that employs the removal of the worst particles to accelerate the swarm, enabling particles to escape local minima and improve the propulsive efficiency of the fish robot. Simulation results demonstrate that the developed M-PSO consumes less energy and requires less time compared to the original PSO and genetic algorithm (GA). Moreover, the M-PSO was tested on a robotic fish navigating an unknown environment characterized by complex spatiotemporal parameters, showcasing its superiority over other methods in all case studies.
Volume: 40
Issue: 1
Page: 10-17
Publish at: 2025-10-01

Hybridized deep learning model with novel recommender for predicting criticality state of patient using MIMIC-IV dataset

10.11591/ijai.v14.i5.pp3926-3933
Sarika Khope , Deepali Kotambkar , Rama Vasantha Adiraju , Smita Suhas Battalwar
The contribution of machine learning towards prediction of critical state of patient is the prime focus of the current study. The review of current approaches of machine learning has been witnessed with various shortcomings. Hence, the proposed study adopts medical information mart for intensive care (MIMIC-IV) dataset in order to develop a novel analytical model that can predict the criticality state of patient in their next visit. The model has been designed by hybridizing convolution neural network (CNN) and long short-term memory (LSTM) which takes the discrete input of hospital and individual patient information in each visit. The concatenated feature is then subjected to a newly introduced recommender module which offers implicit feedback by assigning a ranking score. The final predictive outcome of study offers criticality rank. The study model is benchmarked with existing machine learning approaches to find 54% of increased accuracy and 70% of reduced processing time.
Volume: 14
Issue: 5
Page: 3926-3933
Publish at: 2025-10-01

Development of a smart portable cupping suction device with multi-mode control using PID regulation

10.11591/ijece.v15i5.pp5003-5018
Mohd Riduwan Ghazali , Mohd Ashraf Ahmad , Luqman Hakim Akmalmas
Cupping therapy is a well-established traditional treatment with various health benefits. However, existing electric cupping devices lack precise pressure control and portability which limit their usability across different skin types. This paper presents the development of a smart and portable cupping suction device with multi-mode functionality for dry, wet, and massage cuppings. Designed using an ESP32C3 XIAO microcontroller, a differential pressure sensor (MPX5100DP), and a motor driver (L293D) to enable real-time pressure regulation, the system incorporates a proportional-integral derivative (PID) to maintain a consistent suction performance at the negative pressures of -25, -35, and -45 kPa. The device was tested on different skin conditions of clean, less hairy, and slightly hairy surfaces. A real-time monitoring interface was additionally integrated using a web server to track the variation in pressure. Experimental results demonstrate effectiveness of the PID control system in achieving stable pressure with minimal fluctuations with enhanced user safety and comfort. It advances the medical devices for therapeutic automation by offering a portable, precise, and user-friendly cupping solution.
Volume: 15
Issue: 5
Page: 5003-5018
Publish at: 2025-10-01

Phishing URL prediction – two-phase model using logistic regression and finite state automata

10.11591/ijeecs.v40.i1.pp356-365
Nisha T N , Dhanya Pramod
The human factor in security is more important when they become the carriers of attacks on enterprises. Phishing attacks can be classified as insider attacks when the employees unintentionally participate in the attack propagation. Since complete user training is a myth, enterprises must implement detection tools for phishing attacks on their network perimeters. This research discusses a two-phase model for phishing URL detection, in which the first phase identifies the properties of URLs that detect phishing and their relative weight using logistic regression. The second phase checks the probability of a new URL being categorized as phishing using the knowledge achieved during the first phase using the dynamically created Finite state machines. The model defines a malicious score (MS), which can be used to check any URL in real-time to identify whether it is phishing or not. The model described in this work has been experimented with different benchmarking datasets to verify the performance. The model provided a decent result in classifying a URL as phishing or naive. The malicious score (MS) defined by this model can be used to evaluate any URL and can be used as a filtering mechanism for end-point phishing URL detection. The key contribution is towards developing a two-phase model which evaluates the URL with the help of self-crafted features without reliance on a feature set. This accommodates the model's hyper-competitive phishing URL detection area in cyber security.
Volume: 40
Issue: 1
Page: 356-365
Publish at: 2025-10-01

Machine identification codes of color laser printers: revisiting privacy and security

10.11591/ijeecs.v40.i1.pp137-145
Shreya Arora , Rajendra Kumar Sarin , Pooja Puri
Forging legal documents has been easier and faster with the advancement of technology. Printer identification has become a critical field for tracing criminals and validating the authenticity of documents. The current study uses a non-destructive method to detect and identify covert embedded hidden information (machine identification codes (MIC)). Samples were collected from popular brands, including Xerox and HP color laser printers, to attain this aim. Their printouts were then scanned at 600 dpi using a Konica Minolta scanner. Scanned images were subjected to graphic editors for linear and non-linear adjustments. Following this, yellow-toner dots were observed as a base pattern. Grayscale imaging with a computational approach to analyze the yellow dot patterns was utilized for intensity-focused analysis, with edge detection algorithms applied using Python to enhance and highlight the converted patterns in printed documents. The printouts from Xerox printers exhibited repeating patterns. However, no such detailed information was observed in prints from HP printers, even when analyzed using binary code for deductions. A notable variation was detected in the yellow tracking dots among both brands, which can be instrumental in identifying the origin of printouts and scanned images for forensic investigations. This methodology provides conclusive and dependable accuracy.
Volume: 40
Issue: 1
Page: 137-145
Publish at: 2025-10-01
1 2 3
...
1994 1995
Show 108 of 1995

Discover Our Library

Embark on a journey through our expansive collection of articles and let curiosity lead your path to innovation.

Explore Now
Library 3D Ilustration